artin zur Nedden, HU Berlin 1 Trigger bei Atlas Triggering at ATLAS Trigger Challenge at the LHC Technical Implementation Trigger Strategy, Trigger Menus, Operational Model, Physics Analyses and all that Vortrag von Johannes Haller, Uni HH Am ATLAS-D Meeting, September 2006
Triggering at ATLAS. Vortrag von Johannes Haller, Uni HH Am ATLAS-D Meeting, September 2006. Trigger Challenge at the LHC Technical Implementation Trigger Strategy, Trigger Menus, Operational Model, Physics Analyses and all that. Physics Goals at the LHC. n. -. e. e. q. -. c. 1. ~. - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Martin zur Nedden, HU Berlin 1Trigger bei Atlas
Triggering at ATLAS
Trigger Challenge at the LHC Technical Implementation Trigger Strategy, Trigger Menus,
Operational Model, Physics Analyses and all that
Vortrag von Johannes Haller, Uni HHAm ATLAS-D Meeting, September 2006
Martin zur Nedden, HU Berlin 2Trigger bei Atlas
Physics Goals at the LHC
p pH
µ +
µ-
µ+
µ-
Z
Z
p p
e- e
q
q
q
q
1-
g~
~
20~
q~
10~
EW symmetry breaking ?- search for the Higgs Boson
Extensions of the Standard Model ?- search for SUSY or other BSM physics
p pH
“The” trigger question:
What events do we need to take?
“The” trigger question:
What events do we need to take?
What else? - top, EW, QCD, B-physics
- physics events: , , e, , jets, ET,miss-high pT objects (un-pre-scaled)-low pT objects (pre-scaled or in exclusive selection)
- monitor events- calibration events
- physics events: , , e, , jets, ET,miss-high pT objects (un-pre-scaled)-low pT objects (pre-scaled or in exclusive selection)
• local track finding for LVL1 done on-detector (ASICs)
• looking for coincidences in chamber layers
• programmable widths of 6 coincidence windows determines pT threshold
• looking for coincidences in chamber layers
• programmable widths of 6 coincidence windows determines pT threshold
algorithm:
Available thresholds:
• Muon: 6
Available thresholds:
• Muon: 6
Martin zur Nedden, HU Berlin 12Trigger bei Atlas
LVL1 Trigger Decision in CTP
CTP: (one 9U VME64x crate, FPGA based)central part of LVL1 trigger system
CTP: (one 9U VME64x crate, FPGA based)central part of LVL1 trigger system
signals from LVL1 systems:8-16 EM, 0-8 TAU8 JET, 8 FWDJET4 XE, 4 JE, 4 TE, 6 Muon
signals from LVL1 systems:8-16 EM, 0-8 TAU8 JET, 8 FWDJET4 XE, 4 JE, 4 TE, 6 Muon
internal signals:2 random rates2 pre-scaled clocks8 bunch groups
internal signals:2 random rates2 pre-scaled clocks8 bunch groups
other externalsignals e.g. MB scintillator, …
other externalsignals e.g. MB scintillator, …
calculation of trigger decision for up to 256 trigger items:e.g. “XE70+JET70” raw trigger bits
application of pre-scale factors actual trigger bits
application of veto/ dead time
all of these steps need to be taken into account in offline data analysis all of these steps need to be taken into account in offline data analysis
CTP in USA15:
note: 2 different dead-time settings: trigger groups with “high” and “low” priority will see different luminosities!
CTP
L1A
Martin zur Nedden, HU Berlin 13Trigger bei Atlas
Interface to HLT: RoI Mechanism
LVL1
• triggers on (high) pT objects
• L1Calo and L1Muon send Regions of Interest (RoI) to LVL2 for e//-jet- candidates above thresholds
LVL1
• triggers on (high) pT objects
• L1Calo and L1Muon send Regions of Interest (RoI) to LVL2 for e//-jet- candidates above thresholds
LVL2
• uses Regions of Interest as “seed” for reconstruction (full granularity)
• only data in RoI are used• advantage: total amount of
transfered data is small• ~2% of the total event data• can be dealt with at 75 kHz
LVL2
• uses Regions of Interest as “seed” for reconstruction (full granularity)
• only data in RoI are used• advantage: total amount of
transfered data is small• ~2% of the total event data• can be dealt with at 75 kHz
EF runs after event building, full access to eventEF runs after event building, full access to event
Martin zur Nedden, HU Berlin 14Trigger bei Atlas
ATLAS Trigger & DAQ Architecture
• LVL2 and EF run in large PC farms on the surface
• DAQ and HLT closely coupled• pre-series (corr. ~10% of HLT)
• LVL2 and EF run in large PC farms on the surface
• DAQ and HLT closely coupled• pre-series (corr. ~10% of HLT)
HLT HW:DESY, Humboldt
Martin zur Nedden, HU Berlin 15Trigger bei Atlas
Staging of HLT Components
2006 2007 2008 2009
L2P – LVL2 PC 30 270 510 510
SFI – EventBuilder 32 48 96 128
EFP – EventFilter PC 93 837 1581 1922
SFO – Storage element 3 10 10 10
deferred due to
financial constraints
max LVL1 rate per L2P: 150 Hz
EventBuilder rate per SFI: 40 Hz
max EB rate per EFP: 2 Hz
physics storage rate per EFP: 0.1 Hz
storage rate per storage element: 60 MB/s
40 Hz for 1.5 MB
SFOs non-deferred; allow b/w for calib., debug, etc
max LVL1 rate per L2P: 150 Hz
EventBuilder rate per SFI: 40 Hz
max EB rate per EFP: 2 Hz
physics storage rate per EFP: 0.1 Hz
storage rate per storage element: 60 MB/s
40 Hz for 1.5 MB
SFOs non-deferred; allow b/w for calib., debug, etc
consequences for physics:e.g. in 2007/2008: • LVL1 rate: ~40 KHz (cf. design:75/100 KHz)
• physics storage: ~80 Hz(cf. design: 200 Hz)
consequences for physics:e.g. in 2007/2008: • LVL1 rate: ~40 KHz (cf. design:75/100 KHz)
• physics storage: ~80 Hz(cf. design: 200 Hz)
Martin zur Nedden, HU Berlin 16Trigger bei Atlas
Trigger Strategy
Martin zur Nedden, HU Berlin 17Trigger bei Atlas
HLT Selection Strategy
2) seeded reconstruction• algorithms use results from previous steps
• initial seeds for LVL2 are LVL1 RoIs
2) seeded reconstruction• algorithms use results from previous steps
• initial seeds for LVL2 are LVL1 RoIs
1) step-wise processing and decision• inexpensive (data, time) algorithms first,
complicated algorithms last.
1) step-wise processing and decision• inexpensive (data, time) algorithms first,
complicated algorithms last.
Example: Dielectron Trigger
ATLAS trigger terminology: Trigger chain Trigger signature (called item in LVL1) Trigger element
fundamental principles:
LVL2 confirms & refines LVL1 EF confirms & refines LVL2
note: EF tags accepted events according to physics selection ( streams, offline analysis!)
LVL2 confirms & refines LVL1 EF confirms & refines LVL2
note: EF tags accepted events according to physics selection ( streams, offline analysis!)
Martin zur Nedden, HU Berlin 18Trigger bei Atlas
in parallel: Trigger Chains
in principle: N-Level trigger systembut: Only one pre-scale per chain per level.(to be discussed if used in HLT)
in principle: N-Level trigger systembut: Only one pre-scale per chain per level.(to be discussed if used in HLT)
ATLAS follows “early reject” principle:
- Look at signatures one by one
i.e. do not try to reconstruct full event upfront
if no signatures left, reject event
- Save resources
minimize data transfer and required CPU power
ATLAS follows “early reject” principle:
- Look at signatures one by one
i.e. do not try to reconstruct full event upfront
if no signatures left, reject event
- Save resources
minimize data transfer and required CPU power
HLT Steering enables running of Trigger Chains in parallel w/o interferenceHLT Steering enables running of Trigger Chains in parallel w/o interference
Trigger Chains are independent:
“easy” to calculate trigger efficiencies
“easy” to operate the trigger (finding problems, pre-dictable behavior)
scalable system
Trigger Chains are independent:
“easy” to calculate trigger efficiencies
“easy” to operate the trigger (finding problems, pre-dictable behavior)
scalable system
Martin zur Nedden, HU Berlin 19Trigger bei Atlas
Physics Analysis: the Trigger Part
Every physics analysis needs dedicated thoughts about the trigger:
• trigger rejects 0.999995 more or less hard cuts (in the signal region)
• (each) trigger has an inefficiency that needs to be corrected (turn-on curve)– Similar to offline reconstruction efficiency, but important difference: no retrospective
optimization: “The events are lost forever.”
• trigger optimization (as early as possible)
• trigger data quality during data-taking is crucial
Every physics analysis needs dedicated thoughts about the trigger:
• trigger rejects 0.999995 more or less hard cuts (in the signal region)
• (each) trigger has an inefficiency that needs to be corrected (turn-on curve)– Similar to offline reconstruction efficiency, but important difference: no retrospective
optimization: “The events are lost forever.”
• trigger optimization (as early as possible)
• trigger data quality during data-taking is crucial
studies of this kind are important and are just starting in ATLAS
studies of this kind are important and are just starting in ATLAS
other methods:- di-object samples (J/ Z0,
Z0+jets)- minimum bias and pre-scaled
low-threshold triggers (“bootstrap”)
- orthogonal selections in HLT (ID, muon, calo)
- …
note:- selection bias to be carefully
checked !- trigger efficiency may depend on
physics sample (e.g. electrons in W e and top) investigate in physics groups
other methods:- di-object samples (J/ Z0,
Z0+jets)- minimum bias and pre-scaled
low-threshold triggers (“bootstrap”)
- orthogonal selections in HLT (ID, muon, calo)
- …
note:- selection bias to be carefully
checked !- trigger efficiency may depend on
physics sample (e.g. electrons in W e and top) investigate in physics groups
rec. Z0-peak
trigger effi.
eta
example: possible monitoring of inclusive lepton triggers:
• reconstruct good Z0 candidates offline (triggered by at least one electron trigger)
• Count second electrons fulfilling trigger
example: possible monitoring of inclusive lepton triggers:
• reconstruct good Z0 candidates offline (triggered by at least one electron trigger)
• Count second electrons fulfilling trigger
time-evolution of accuracy
number of events
total efficiency for muons
electronpositron
Martin zur Nedden, HU Berlin 22Trigger bei Atlas
LVL1 Menu (as of today, TDR)
general trigger problem: cover as much as possible of the kinematic phase space for
physics low trigger thresholds
keep the trigger rate low high trigger thresholds
trigger menu is a compromise
general trigger problem: cover as much as possible of the kinematic phase space for
physics low trigger thresholds
keep the trigger rate low high trigger thresholds
trigger menu is a compromise LVL1 rate is dominated by electromagnetic clusters: 78% of physics triggers
LVL1 rate is dominated by electromagnetic clusters: 78% of physics triggers
Note:
• large uncertainties on predicted rates
• study of the global aspects needed: load balancing (e.g. jet triggers)
Note:
• large uncertainties on predicted rates
• study of the global aspects needed: load balancing (e.g. jet triggers)
Martin zur Nedden, HU Berlin 23Trigger bei Atlas
HLT Menu (as of today, TDR)
e/ rate reduced mainly in LVL2 (full granularity in RoI)
e/ rate reduced mainly in LVL2 (full granularity in RoI)
Note:
• large uncertainties on predicted rates (no data!)
• these menu give an rough impression of what we will select.
• details of the menu are not yet worked out (pre-scales, monitoring, …)
• but first examples of realistic trigger menus needed soon
Note:
• large uncertainties on predicted rates (no data!)
• these menu give an rough impression of what we will select.
• details of the menu are not yet worked out (pre-scales, monitoring, …)
• but first examples of realistic trigger menus needed soon
Martin zur Nedden, HU Berlin 24Trigger bei Atlas
towards a more complete Menu
• study slice-wise:- optimization of cuts- need distributions of rates, rate vs. eff- more realism to algorithms- detailed studies of threshold behaviour, noise- consequences on physics reach
• study of the global aspects: - load balancing (e.g. jet triggers balancing)- overlap between selections, optimization
• the important details of the menu- monitoring strategy- pre-scaling strategy (dynamic, static) triggers- concurrent data-taking (pre-scales) or
sequentially (i.e. dedicated runs)?- time evolution (luminosity, background, etc.)
pre-scale changes ala H1/CDF?- technical triggers (bunch-groups, etc.)
- …
• study slice-wise:- optimization of cuts- need distributions of rates, rate vs. eff- more realism to algorithms- detailed studies of threshold behaviour, noise- consequences on physics reach
• study of the global aspects: - load balancing (e.g. jet triggers balancing)- overlap between selections, optimization
• the important details of the menu- monitoring strategy- pre-scaling strategy (dynamic, static) triggers- concurrent data-taking (pre-scales) or
sequentially (i.e. dedicated runs)?- time evolution (luminosity, background, etc.)
pre-scale changes ala H1/CDF?- technical triggers (bunch-groups, etc.)
- …
-…
aim: get concrete examples of more complete and realistic trigger menus for discussion at the next trigger and physics weeks.
aim: get concrete examples of more complete and realistic trigger menus for discussion at the next trigger and physics weeks.
ad-hoc-group: • started rethinking about the
trigger menus• invites input from physics,
combined performance and detector groups
ad-hoc-group: • started rethinking about the
trigger menus• invites input from physics,
combined performance and detector groups
priorities:• consolidate work on menu
for 14 TeV and 1031.• in parallel: limited study for
0.9 TeV and 1029
• later look at 1032 and above
priorities:• consolidate work on menu
for 14 TeV and 1031.• in parallel: limited study for
0.9 TeV and 1029
• later look at 1032 and above
Martin zur Nedden, HU Berlin 25Trigger bei Atlas
Ideas for early Data Taking
conditions of early data-taking:initial luminosity: 1031(1029), bunch spacing 75ns (~500ns) BCID not critical, can relax the trigger timing windows
conditions of early data-taking:initial luminosity: 1031(1029), bunch spacing 75ns (~500ns) BCID not critical, can relax the trigger timing windows
trigger commissioningunderstanding of LVL1 is crucial at startup
first phase: • rates are low • DAQ can stand 400 MB/s• LVL1 only, HLT “transparent” • some pre-scaling needed only for very low thresholds.
• HLT selections studied offlinesecond phase:
• insert HLT • start with very simple and basic algorithms
trigger commissioningunderstanding of LVL1 is crucial at startup
first phase: • rates are low • DAQ can stand 400 MB/s• LVL1 only, HLT “transparent” • some pre-scaling needed only for very low thresholds.
• HLT selections studied offlinesecond phase:
• insert HLT • start with very simple and basic algorithms
minimum bias events:• important esp. at the beginning:
- crucial for timing-in of the experiment - for commissioning of detectors/ trigger/ offline
selection- physics: as bkg. (important for 14 TeV), per se
• possible implementation:• BC LVL1 trigger + selection on LVL2/EF
• bias free at LVL1
• MBTS trigger at LVL1 + selection in HLT• some bias at LVL1 (η range; efficiency for MIPS;
multiplicity requirements; etc.)• needed where interactions per BC << 1
minimum bias events:• important esp. at the beginning:
- crucial for timing-in of the experiment - for commissioning of detectors/ trigger/ offline
selection- physics: as bkg. (important for 14 TeV), per se
• possible implementation:• BC LVL1 trigger + selection on LVL2/EF
• bias free at LVL1
• MBTS trigger at LVL1 + selection in HLT• some bias at LVL1 (η range; efficiency for MIPS;
multiplicity requirements; etc.)• needed where interactions per BC << 1
Martin zur Nedden, HU Berlin 26Trigger bei Atlas
The technical Side: Trigger Configuration
LVL1 HLT
Unique key
•TrigConf system under development•real data-taking: trigger menu can change between runs
optimization,falling luminosity during a fill (pre-scales, cuts)…
•book-keeping of all settings crucial
•TrigConf system under development•real data-taking: trigger menu can change between runs
optimization,falling luminosity during a fill (pre-scales, cuts)…
•book-keeping of all settings crucial
Offline data analyzer users will have to look up the TriggerDB to interpret the trigger result in the events, e.g. to find the settings for their triggers and the corresponding run ranges.
Offline data analyzer users will have to look up the TriggerDB to interpret the trigger result in the events, e.g. to find the settings for their triggers and the corresponding run ranges.
•TriggerDB is central part:stores all information for the online selectionstores all versions of trigger settings.identified with a unique key to be stored in
CondDB.
•TriggerDB is central part:stores all information for the online selectionstores all versions of trigger settings.identified with a unique key to be stored in
CondDB.
Martin zur Nedden, HU Berlin 27Trigger bei Atlas
The technical side: Trigger Configuration
Java front-end for the TriggerDB under development: TriggerTool
• three modes are foreseen:1) experts: construct consistent menus in
TriggerDB2) shift-crew: choice of predefined options
(menus, pre-scale sets)
3) offline user: extract menus in text file for development, or simulation etc, browse DB to find settings of triggers and run ranges
Java front-end for the TriggerDB under development: TriggerTool
• three modes are foreseen:1) experts: construct consistent menus in
TriggerDB2) shift-crew: choice of predefined options
(menus, pre-scale sets)
3) offline user: extract menus in text file for development, or simulation etc, browse DB to find settings of triggers and run ranges